KVR Audio

kmonkey · Post by **kmonkey** » Sun Oct 31, 2010 5:39 pm

Some serious software-synth hacking in this thread

sonicpowa · Post by **sonicpowa** » Sun Oct 31, 2010 5:49 pm

djsubject wrote:
Howard wrote:
djsubject wrote:re-listened and 1 still sounds the best to me!!
its stronger than 2, i love it!!
Agree, #1 sounds better.
well, thats helped me to trust my ears a bit more

looking forward to seeing in what form this experiment will reach us,

Subz

Well, 'better' is a bit subjective.
#1 is more 'hifi' and bigger sounding but #2 is still more interesting. If you listen carefully you'll hear a beautiful juicy chorus-like sound but it's not smeared like a typical chorus.
#1 can be achieved with pretty much anything, especially with Sylenth1.
(IMO of course)
edit: also #1 has that famous 'digitalis' quality.

Music Engineer · Post by **Music Engineer** » Sun Oct 31, 2010 5:59 pm

aciddose wrote: i'm pretty certain it can't be exactly the maximal LCM because if it were, then those would be fractions of primes, right? (being no LCM in that case)

not sure about the primes thing, but however, when we raise some number to the power of 0.5, i.e. we take the square root, the result will likely to be irrational in which case there actually won't be a finite exact LCM anyway (i think). but we also have to worry about approximate alignments corresponding to approximate LCMs etc. but perhaps it's safe to say that we want to minimize the occurences of (approximately) phase aligned saws

aciddose · Post by **aciddose** » Sun Oct 31, 2010 6:04 pm

also - it doesn't need to be square root, it can be any value. i find from 1/8 to 8 to be a good range. the timbres change but apparently it maintains the lack of correlation.

ah ha!

"Square roots of integers that are not perfect squares are always irrational numbers: numbers not expressible as a ratio of two integers"

Music Engineer · Post by **Music Engineer** » Sun Oct 31, 2010 6:06 pm

aciddose wrote:also - it doesn't need to be square root, it can be any value. i find from 1/8 to 8 to be a good range. the timbres change but apparently it maintains the lack of correlation.

good to know. perhaps this could be even nice for a user parameter

aciddose · Post by **aciddose** » Sun Oct 31, 2010 6:10 pm

it is, i make it available in xhip. it changes the timbre from being "wide" and "smooth" near 0.5 to "thin" and "digital" near 4.0, with the extremes exactly that - more extreme.

i use the same distribution in my reverb http://xhip.presetexchange.com/temp/ad_reverb.dll where you can see the "distribution" parameter. i'm not certain i have this implemented correctly though, i just re-wrote the thing a few days ago. anyway the error is i think to do with offsets, it seems to still have the low-correlation effect as expected.

Music Engineer · Post by **Music Engineer** » Sun Oct 31, 2010 6:18 pm

aciddose wrote:it is, i make it available in xhip. it changes the timbre from being "wide" and "smooth" near 0.5 to "thin" and "digital" near 4.0, with the extremes exactly that - more extreme.

and to the other side, below 0.5?

i use the same distribution in my reverb

for delayline lengths in an FDN?

aciddose · Post by **aciddose** » Sun Oct 31, 2010 6:22 pm

no, it's a unique method which i don't think is documented anywhere. i invented it myself, whether or not it's a reinvention. it uses the delay as a memory for an allpass filter, so it's a series nth order allpass filters on mode-2, mode-1 is a standard combs+allpass configuration. it depends upon both the allpass configuration _and_ delay lengths for it's smooth sound. one without the other doesn't sound very good.

below 0.5 it seems to sound less smooth yet, still uncorrelated. that's actually a fairly interesting effect that i've never really studied.

the global unison in xhip supports these parameters so if you wanted to very quickly fool around with it, run a plain ramp and set the unison to however many voices, then adjust away.

antto · Post by **antto** » Sun Oct 31, 2010 6:24 pm

so, what is this power distribution thing? i seem to missed it

aciddose · Post by **aciddose** » Sun Oct 31, 2010 6:40 pm

didn't we go off topic enough

it's in the thread!

Music Engineer · Post by **Music Engineer** » Sun Oct 31, 2010 6:45 pm

aciddose wrote:below 0.5 it seems to sound less smooth yet, still uncorrelated. that's actually a fairly interesting effect that i've never really studied.

the global unison in xhip supports these parameters so if you wanted to very quickly fool around with it, run a plain ramp and set the unison to however many voices, then adjust away.

mmhh... i don't see it. or do i need the other ("stable") version?

Music Engineer · Post by **Music Engineer** » Sun Oct 31, 2010 6:51 pm

so, what is this power distribution thing? i seem to missed it

page 3:

for fraction 0 to 1 in n steps
result[n] = pow(12throot2, detune depth * pow(fraction, distribution))
result[] -= (average(result[]) - 1) / 2;

aciddose · Post by **aciddose** » Sun Oct 31, 2010 6:58 pm

you need the "stable" version, as i said the one i'm working on currently hasn't got the complete gui.

Music Engineer · Post by **Music Engineer** » Sun Oct 31, 2010 7:15 pm

aciddose wrote:you need the "stable" version, as i said the one i'm working on currently hasn't got the complete gui.

ah, OK, got it. one thing: the first note after adjusting the number of voices seems to start with phase-aligned voices leading to some weird transient. are your oscs generally free-runing but initialize phases on changing the number of unison voices?

Ingonator · Post by **Ingonator** » Sun Oct 31, 2010 7:30 pm

A bit off topic but i guess someone here could answer this:

I just sampled some "Hypersaw" sounds which i made with my "HyperOSC" synth for AAS Tassman 4.

By an basic Google search ("Hypersaw samples") i just found some Hypersaw samples of the Virus on the web.

Could this be legal or only if i don't do a commercial soundbank with it???

Supersaw!